Multi agent systems via  intermittent communications

This study investigates consensus problems for multi-agent systems with second-order dynamics under delayed and intermittent communication. A class of distributed control algorithms are proposed based on the relatively delayed and periodical intermittent information of neighbouring agents. Then, by using tools from Lyapunov approach and graph theory, it is proved that under intermittent and delayed communication, consensus can still be asymptotically achieved if some sufficient conditions are satisfied.

Moreover, the relationship between time delays and communication duration over each control period is sought out. As an extension, consensus tracking problems are studied under a directedcommunication topology containing a spanning tree. Both the cases of partly and completely intermittent information exchange are dealt with. Finally, several simulation examples are provided to illustrate the effectiveness of the proposed theoretical analyses.